Talk:GVCS Presentation Transcript - Revision history

LucasG at 19:00, 26 December 2010

2010-12-26T19:00:15Z

← Older revision		Revision as of 19:00, 26 December 2010
Line 53:		Line 53:
	* Then I've used "ffmpeg -i input.mov -acodec libmp3lame -ac 2 -ab 131072 -vn -y output.mp3" to extract the sound, so I can work in a computer without much room (an acer aspire one). Thanks @ezanardi!		* Then I've used "ffmpeg -i input.mov -acodec libmp3lame -ac 2 -ab 131072 -vn -y output.mp3" to extract the sound, so I can work in a computer without much room (an acer aspire one). Thanks @ezanardi!
	* Next I need a way to listen to 20-second-sized chunks, several times until it's transcribed; then next chunk (or maybe previous if needed) until done.		* Next I need a way to listen to 20-second-sized chunks, several times until it's transcribed; then next chunk (or maybe previous if needed) until done.
	** ezanardi to the rescue:		** ezanardi to the rescue, if you have a bash shell:
	** ini=0 ; while [ true ] ; do mpg123 -v --skip $ini --frames 760 --loop -1 -C input.mp3 ; ini=$(( $ini + 760 )) ; done		** ini=0 ; while [ true ] ; do mpg123 -v --skip $ini --frames 760 --loop -1 -C input.mp3 ; ini=$(( $ini + 760 )) ; done
	** y pasas de un segmento a otro pulsando "q". También puedes repetir un segmento antes de que termine, pulsando "f"		** y pasas de un segmento a otro pulsando "q". También puedes repetir un segmento antes de que termine, pulsando "f"
	** Un mp3 tiene unos 38 frames por segundo. Así que 760 frames cada 20 segundos.		** Un mp3 tiene unos 38 frames por segundo. Así que 760 frames cada 20 segundos.
	* As you see, this is part of helping @universalsubs create a general purpose tool for cooperative transcriptions and translations.		* As you see, this is part of helping @universalsubs create a general purpose tool for cooperative transcriptions and translations.

LucasG at 18:40, 26 December 2010

2010-12-26T18:40:20Z

← Older revision		Revision as of 18:40, 26 December 2010
Line 53:		Line 53:
	* Then I've used "ffmpeg -i input.mov -acodec libmp3lame -ac 2 -ab 131072 -vn -y output.mp3" to extract the sound, so I can work in a computer without much room (an acer aspire one). Thanks @ezanardi!		* Then I've used "ffmpeg -i input.mov -acodec libmp3lame -ac 2 -ab 131072 -vn -y output.mp3" to extract the sound, so I can work in a computer without much room (an acer aspire one). Thanks @ezanardi!
	* Next I need a way to listen to 20-second-sized chunks, several times until it's transcribed; then next chunk (or maybe previous if needed) until done.		* Next I need a way to listen to 20-second-sized chunks, several times until it's transcribed; then next chunk (or maybe previous if needed) until done.
			** ezanardi to the rescue:
			** ini=0 ; while [ true ] ; do mpg123 -v --skip $ini --frames 760 --loop -1 -C input.mp3 ; ini=$(( $ini + 760 )) ; done
			** y pasas de un segmento a otro pulsando "q". También puedes repetir un segmento antes de que termine, pulsando "f"
			** Un mp3 tiene unos 38 frames por segundo. Así que 760 frames cada 20 segundos.
	* As you see, this is part of helping @universalsubs create a general purpose tool for cooperative transcriptions and translations.		* As you see, this is part of helping @universalsubs create a general purpose tool for cooperative transcriptions and translations.

LucasG at 18:31, 26 December 2010

2010-12-26T18:31:17Z

← Older revision		Revision as of 18:31, 26 December 2010
Line 48:		Line 48:

	(This http://openfarmtech.org/wiki/Global_Village_Construction_Set_2_Minute_Explainer_Video is another page I've been working at. Useful for youtube videos.)		(This http://openfarmtech.org/wiki/Global_Village_Construction_Set_2_Minute_Explainer_Video is another page I've been working at. Useful for youtube videos.)

			(lucas)
			* I've downloaded the video from vimeo.
			* Then I've used "ffmpeg -i input.mov -acodec libmp3lame -ac 2 -ab 131072 -vn -y output.mp3" to extract the sound, so I can work in a computer without much room (an acer aspire one). Thanks @ezanardi!
			* Next I need a way to listen to 20-second-sized chunks, several times until it's transcribed; then next chunk (or maybe previous if needed) until done.
			* As you see, this is part of helping @universalsubs create a general purpose tool for cooperative transcriptions and translations.

LucasG at 16:24, 26 December 2010

2010-12-26T16:24:11Z

← Older revision	Revision as of 16:24, 26 December 2010
Line 1:	Line 1:
		(lucas)
		* http://universalsubtitles.org/videos/kGGVxuvDZLVR/ doesn't work for me right now. Original video http://vimeo.com/13020225
		* In my experience, we need to have a good transcript before moving it to the subtitles' platform.
		* This page looks funny, but once you enter edit mode, the lines are unfolded and much nicer to the eye. [[User:LucasG\|LucasG]] 14:25, 9 December 2010 (PST)

		(paolo)
		* I will start to create subtitle only after revision because save a partial work is not easy in universalsubtitle for what I've tryed to do. -paolo

		(lucas)
		* ok, so transcript first! And I've moved this conversation to the front page. We'll move it back to the "talk" page once we're done.
		* Plan:
		** Review the transcript so that it's all good.
		** Cut it in short pieces of text.
		** Place the transcript in universal subtitles.
		** Translate into whatever languages.
		** If someone speaks out the translation, simply following the subtitles with the audio muted and speaking into a recorder, and sends an audio file, we can do overdubbing. (I've tried with the 2-minute video on GVCS and it worked well. Building on each other's work as usual.)

		(paolo)
		I've just finish to write it. There's a lot of holes and I think a lot of mistakes. I've put a --- where I wasn't able to write something near what marcin was saying.
		Can someone finish it? I can recheck it but I'm not able to finish it, there's some idiomatic form that I don't know and understand.
		Chose what do you want I do please. thank for help! I will make subtitle and translate them into Italian then!

		(lucas)
		Thanks Paolo.
		* I've downloaded the video and intend to finish before the week's end:
		** my revision (other reviewers welcome, of course!),
		** then cutting it up in small chunks along with spoken words,
		** and finally placing the text
		*** both in http://www.universalsubtitles.org
		*** and for google-aided translation http://translate.google.com/#en\|es\|
		* Then we have a few days before the year's end to work in parallel, and see how fast we can go.

		(paolo)
		Can somebody explain in short to me what translate and universalsubtitle support?
		Is more simple use a program for subtitling out the internet and then upload the file to universal subtitle and google translate??

		(lucas)
		* We'll do the plain text transcription first. With each sentence in its own line. If some lines are long, then we cut them in two or more pieces.
		* After that, we need to convert that flat text into a .srt file = with timing included. Just like you can see here: http://openfarmtech.org/wiki/Video_translation
		** I have used this tool: "divxland media subtitler 2.0.7 for windows", which is free as in money, and easy to use: 1) add text, 2) time done pressing button as needed. http://www.divxland.org/subtitler.php
		** But I think we can use universalsubtitles which works the same way.
		** I think we just need to try.
		* Once we have a well timed .srt file, we can:
		** upload it to universalsubs if it wasn't already there.
		** Place it in google translate, which lets you have 2 columns, one with the original language and one with the new language. I found it very easy to use. We create a new .srt which we can upload to universalsubs.

		I think once this is done I'll be able to write down the whole process.

		(This http://openfarmtech.org/wiki/Global_Village_Construction_Set_2_Minute_Explainer_Video is another page I've been working at. Useful for youtube videos.)

LucasG: Blanked the page

2010-12-11T08:54:16Z

Blanked the page

← Older revision		Revision as of 08:54, 11 December 2010
Line 1:		Line 1:
	~~I will start to create subtitle only after revision because save a partial work is not easy in universalsubtitle for what I've tryed to do. -paolo~~

Pignol23: Created page with "I will start to create subtitle only after revision because save a partial work is not easy in universalsubtitle for what I've tryed to do. -paolo"

2010-12-11T02:10:24Z

Created page with "I will start to create subtitle only after revision because save a partial work is not easy in universalsubtitle for what I've tryed to do. -paolo"

New page

I will start to create subtitle only after revision because save a partial work is not easy in universalsubtitle for what I've tryed to do. -paolo